LyDROO: Adaptive Computation Offloading in MEC Using Deep Reinforcement Learning

DOI

10.2991/978-94-6463-940-7_31How to use a DOI?

Keywords

Mobile Edge Computing (MEC); Deep Reinforcement Learning (DRL); Lyapunov Optimization; Computation Offloading; LyDROO

Abstract

Mobile Edge Computing (MEC) brings cloud-like capabilities closer to end users, enabling low-latency and high-efficiency processing for applications like autonomous vehicles, virtual reality, and smart healthcare. A core challenge in MEC is adaptive computation offloading, deciding whether tasks should be processed locally or offloaded to edge servers while considering energy consumption, network dynamics, and system stability. Traditional rule-based and optimization-based methods are often rigid or computationally intensive, and while Deep Reinforcement Learning (DRL) offers adaptability, it struggles with stability, convergence speed, and power efficiency. To address these challenges, we propose LyDROO, a hybrid framework that integrates Lyapunov optimization with Deep Reinforcement Learning. LyDROO decomposes long-term offloading objectives into short-term, solvable decisions and leverages neural networks (DNNs and CNNs) to learn optimal strategies dynamically. Simulation results demonstrate that LyDROO significantly improves task completion time, energy efficiency, and queue stability, outperforming existing DRL and optimization methods. The framework shows strong adaptability and scalability, making it suitable for next-generation intelligent MEC systems.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - V. Srinivas Lokavarapu
AU  - Kunjam Nageswara Rao
AU  - Shiva Shankar Reddy
AU  - Sitaratnam Gokuruboyina
PY  - 2025
DA  - 2025/12/31
TI  - LyDROO: Adaptive Computation Offloading in MEC Using Deep Reinforcement Learning
BT  - Proceedings of the Conference on Social and Sustainable Innovation in Technology & Engineering (SASI-ITE 2025)
PB  - Atlantis Press
SP  - 421
EP  - 432
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-940-7_31
DO  - 10.2991/978-94-6463-940-7_31
ID  - Lokavarapu2025
ER  -